Facial recognition may be utilized as a security feature for an electronic device connected to or equipped with a camera. For example, a user may be asked to submit a face capture during the initial setup of a device. During the setup or enrollment procedure, the user's face may be captured and stored as an authorized user image. A facial recognition analysis may be performed on the captured image to designate features of the user's face and/or determine a computer model of the user's face (or derivatives thereof). Subsequently, if an individual seeks access to the device, a picture of the individual may be captured, analyzed, and compared to the stored authorized user's image (and corresponding analysis) to determine if the individual matches that image. In the event insufficient correlation exists between the image from the individual attempting to access the device and the authorized user's image, access to the device may be denied.
Despite the above verification method, an unauthorized user may attempt to use a spoofing method to gain access to the device, such as using a printed pictured of the authorized user. One method to detect a picture may use a proximity sensor (e.g., laser range finder, infrared proximity technology, photometric technology, radar, sonar, etc.) may measure the distance between a computing device and an object external to the device (e.g., a user or picture of the user). The proximity information may be used to determine an expected number of pixels between two features on a captured image (e.g., distance between pupils). A user may be denied access to the device if the number of pixels in an access request capture picture exceeds a threshold value that is expected based on the distance to the object and the authorized image.